Skip to content

Pull requests: intel/intel-xpu-backend-for-triton

New pull request New
48 Open 4,767 Closed
48 Open 4,767 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

[vLLM][MoE] Avoid top-k boundary ties causing test failures
#7580 opened Jul 24, 2026 by sarah12121212 Contributor • Draft
[TritonIntelGPUToLLVM] Vectorize masked StoreOp when mask isn't statically uniform run-benchmarks
#7577 opened Jul 24, 2026 by anmyachev Contributor • Draft
1
2
Fix tt.atomic_poll runtime crash for 16-bit types
#7573 opened Jul 24, 2026 by wdziurdz Contributor Loading…
1
1
Merge OpenAI Triton commit c393f1b
#7572 opened Jul 24, 2026 by wdziurdz Contributor Loading…
[AxisInfo] Fix signed division/remainder analysis and revert PyTorch floordiv workaround
#7570 opened Jul 23, 2026 by whitneywhtsang Contributor • Draft
1
[vLLM][moe] Route blockwise FP8 fused MoE to Triton
#7569 opened Jul 23, 2026 by sarah12121212 Contributor • 2/2 • Draft
Remove deprecated block pointer from batched flash attention benchmark run-benchmarks
#7568 opened Jul 23, 2026 by whitneywhtsang Contributor • Draft
[vLLM] Add vLLM BMG benchmarks to run-benchmarks label and benchmarks/ source change trigger
#7567 opened Jul 23, 2026 by quinnlp Contributor • Draft
[vLLM] Move vLLM test schedules into per-arch wrapper workflows
#7560 opened Jul 22, 2026 by quinnlp Contributor • Draft
[vLLM] Add run-vllm-benchmarks / run-vllm-tests PR labels run-vllm-benchmarks run-vllm-tests
#7559 opened Jul 22, 2026 by quinnlp Contributor • Draft
[CI] Build on Windows: use GitHub-hosted windows-2022 CPU runner (#7443)
#7555 opened Jul 22, 2026 by slawblauciak Contributor Loading…
1
3
Improve codegen retry for CRI.
#7551 opened Jul 22, 2026 by chengjunlu Contributor Loading…
1
25
Add triton intel gpu gather op lowering.
#7548 opened Jul 22, 2026 by chengjunlu Contributor • Draft
1
[vLLM] Update vLLM pin
#7539 opened Jul 20, 2026 by quinnlp Contributor • Draft
1
artur fused moe tdesc
#7518 opened Jul 17, 2026 by quinnlp Contributor • Draft
Support to load transpose B matrix of sub-group-size=32.
#7488 opened Jul 16, 2026 by Copilot AI • Draft
@chengjunlu
8
[vLLM] [unified-attention] Blockm autotune prune logic fix
#7476 opened Jul 15, 2026 by sarah12121212 Contributor • Draft
1
[vLLM][MoE] Fix WorkspaceManager initialization for vLLM MoE tests
#7462 opened Jul 14, 2026 by sarah12121212 Contributor • 1/2 • Draft
[vLLM]Skip Marlin MoE tests
#7461 opened Jul 14, 2026 by sarah12121212 Contributor • Draft
[vLLM][moe] Skip unsupported XPU WNA16 MoE quantization modes and update skiplist
#7452 opened Jul 14, 2026 by sarah12121212 Contributor • Draft
[BlockIO] Use 2D block load for 32-bit and wider descriptor row lengths
#7444 opened Jul 14, 2026 by wdziurdz Contributor Loading…
1
[vLLM][moe] Keep batched moe reference activation eager, guard CUDA stream-capture checks, and update skiplist
#7440 opened Jul 14, 2026 by sarah12121212 Contributor • Draft
[vLLM][moe] Enable DeepGEMM SiLU FP8 moe fallback and update skiplist
#7439 opened Jul 14, 2026 by sarah12121212 Contributor • Draft
[Debug] Proton "sampling" mode
#7412 opened Jul 10, 2026 by anmyachev Contributor • Draft
[vLLM][unified_attention] Add TILE_SIZE autotuning
#7405 opened Jul 8, 2026 by sarah12121212 Contributor • Draft
ProTip! Updated in the last three days: updated:>2026-07-21.